Previous studies (Urenjak, et al. 1993; Floran, 1995) have shown that the metabolic profiles revealed by proton NMR spectroscopy can be useful to distinguish unambiguously between different brain cell populations. In addition, 1H-NHR spectra could similarly be used to distinguish between different types of tumors of the human nervous system and to relate tumor populations to their cell type of origin. Preliminary data suggest that these spectra may enable to identification of metabolic changes associated with conditional oncogene activation, and that certain metabolite differences may allow normal brain cells to be distinguished from their transformed counterparts. This would allow for the non-invasive, non-surgical diagnosis of human CNS tumors. The research for my dissertation will use 1H-NMR spectroscopy in vitro, in order to compare normal and tumor cells of the same lineage in order to test the hypothesis that 1H-NHR spectroscopy can be used both to determine the lineage of a tumor cell an to investigate metabolic alterations that may distinguish tumor cells from normal cells. In addition, my research will include 1H-NMR spectroscopy of superfused cells ex vivo, in order to relate spectra from the in vitro cell extracts to the in vivo situation of observing metabolizing cells. Finally, my research will include 1H-NMR spectroscopy of implanted tumor cells in the rat brain in vivo, in order to obtain information that would aid the non- invasive diagnosis by NMR spectroscopy of various types of human brain tumors, in terms of their histological origin and degree of malignancy. The ultimate goal of this research is the identification and regulation of the pathways which may be differentially activated in tumor cells, and of the specific metabolic changes characteristic of the transformation process of various cell types of the CNS.